repricing — what changed price in the x402 Bazaar and what happened afterwards

1 · price changes between the two latest snapshots

6 listings repriced4 sellers4 raised2 lowered2 earlier changes with a 7-day reading

Prices in USDC per call. A sub-table under a listing is section 2: that listing's earlier price changes and what the demand readings did around them. The last column is the exact request body for the paid seller report on that listing.

bykaranteli.com · payTo 0xb4C6c903DeE310e4060d8E6eB68582426bE1baa9 · 3 listings (3 raised, 0 lowered)

mcp.whitemagic.dev · payTo 0x213b6bB4fB073E86167c58DA6Cb6E9477557DaA8 · 1 listing (0 raised, 1 lowered)

api.seerium.xyz · payTo 0x81832eEDe3a4ad96CB3D7cA09977636D912fd3D5 · 1 listing (1 raised, 0 lowered)

dripstack.com · payTo 0xBF22b6DdB5A08c823856A779f1004eEa60C5aB92 · 1 listing (0 raised, 1 lowered)

2 · earlier changes, read

For each listing above, every earlier price change with at least 7 days of history on both sides is shown under it: payers and calls (trailing 30 days) on the last day at the old price and exactly 7 days after the change, and the revenue the calls observed at the new price would have produced at the abandoned price. These are the same readings the paid report returns, from the same code.

3 · rolling 30 days

906 repricing events823 listings473 raised433 lowered

581 events with a 7-day readingpayers_30d fell after 11 of them (2%)rose after 297unchanged after 273325 too recent to read

2026-09-04 → 2026-10-03. an event has a 7-day reading when the listing was still priced 7 days after the change; payers_30d there is compared with the previous snapshot's.

4 · how these numbers are derived

Every day the observatory takes one snapshot of the x402 Bazaar catalog: for each listing, the price its 402 terms publish (amount_atomic, USDC with six decimals) and the trailing 30-day call and payer counts the catalog reports for it. Nothing here is a seller self-report to us; it is what the listing itself publishes.

A repricing is a snapshot day on which a listing's recorded price differs from the previous snapshot's. One observation per day means a price that changed and changed back within a day is not seen, so counts here are conservative.

For an earlier price change, "day before" is the last snapshot at the old price and "7 days after" is the snapshot exactly seven days after the first snapshot at the new price. Both are trailing 30-day counts, so a 7-day reading carries 23 days of pre-change history in it; the direction of the move is the signal, not its size.

The counterfactual takes the calls observed at the new price (the last snapshot at which that price still held) and asks what the abandoned price would have earned on that same volume. regret_usdc positive means the abandoned price would have earned more; negative means the reprice earned more. No elasticity is inferred beyond the observed snapshots.

These readings use the same functions as the paid seller report, so a number on this page and the number in a paid call for the same listing never disagree.

your own listing

Free worked example (a real listing that repriced several times, with the input and response schemas): https://api.oblique.markets/api/v1/reprice-regret/example

Paid seller report, $0.01 per call, x402 (USDC on Base or Solana) or MPP (pathUSD on Tempo): POST https://api.oblique.markets/api/v1/paid/reprice-regret

{"resource_url":"https://<host>/<path exactly as listed>","window_days":30}
{"seller_payto":"<the payTo address the listing publishes>","window_days":30}
# exactly one of resource_url or seller_payto; window_days 7–90 (default 30); optional network, cursor

Call it without payment first: the HTTP 402 answer carries the exact amount, asset and payTo; pay and retry with the payment header. No account, no key.